Solved: Confused About Facebook ABO vs CBO Campaigns

Hi there,

Thanks for reaching out! Read through your situation and it's a really common one, especially for folks coming over from Google Ads. Meta can feel like a total black box when you're used to the intent-based world of search. I'm happy to give you some initial thoughts and guidance on how to get out of this creative testing loop and build something more stable and scalable.

Honestly, the issue isn't really about ABO vs CBO. That's a symptom, not the cause. The real problem is the process you're using to test and the structure of your account. Let's get that sorted.

We'll need to look at why your creatives aren't getting spend...

Okay, first thing's first. You need to stop uploading three new creatives every single day. I know it feels productive, and tools like the one you're using make it easy, but you're actually sabotaging your own account. It's the single biggest reason you're feeling frustrated and why your new ads aren't getting any budget.

Think of the Meta algorithm like a new employee you've just hired. On day one, you give them three tasks. Before they can even figure out the best way to do the first one, you interrupt them and give them three more completely different tasks. Then you do it again the next day, and the next. That employee would never learn how to do any single task well, and they'd probably end up just doing a tiny bit of the one task that looked easiest at the start. That's exactly what you're doing to the Meta algorithm.

Every time you add a new creative to an ad set, it triggers what's called the "Learning Phase". During this phase, the algorithm needs to show the ad to a bunch of different people to figure out who responds best to it. It needs around 50 conversions (purchases, leads, etc.) within a 7-day period to exit this phase and start performing optimally. By adding new stuff daily, your ad sets are in a permanent state of learning. They never gather enough data on any single creative to make a smart decision, so the CBO does the only logical thing it can: it pushes all the budget to the one or two creatives that got a bit of early, lucky traction and ignores the rest. It's not broken; it's doing exactly what it's designed to do in the chaotic environment you've created for it.

A ROAS of 3.5 is actually not bad at all, it's a pretty solid foundation. It tells me your product and your offer are decent. But the way you're managing the account means that 3.5 is likely very unstable and unpredictable. One day it's 5, the next it's 1.5. Our goal isn't just to get a good ROAS, it's to build a system that produces a predictable ROAS, day in and day out, so you can scale your £700/day spend to £1000, then £2000, and beyond, with confidence.

The feeling of it being a "black box" compared to Google is completely understandable. On Google, you're catching fish in a barrel. Someone types in "buy red handcrafted widgets," you show them an ad for red handcrafted widgets. The intent is already there. On Meta, you're the fisherman on the boat trying to attract fish that weren't even looking for a meal. You have to interrupt their day (scrolling through photos of friends and memes) with a message so compelling it makes them stop and listen. This requires a completely different approach to structure and testing.

I'd say you should stop thinking about ABO vs CBO...

The whole "ABO vs CBO" debate is one of the biggest distractions in paid advertising. One isn't inherently 'better' than the other; they are just different tools for different jobs. Arguing about which is better is like a carpenter arguing whether a hammer is better than a screwdriver. It's a pointless conversation without context.

• CBO (Campaign Budget Optimisation) is for scaling. You give it a pile of money, a bunch of proven audiences, and proven creatives, and you tell it, "You're the expert, you figure out the most efficient way to spend this money to get me the cheapest conversions." It's brilliant at optimising between known good assets.
• ABO (Ad Set Budget Optimisation) is for testing and control. You set a specific budget at the ad set level, and you're telling the algorithm, "I don't care what you think is best. I want you to spend exactly this much money on this specific audience so I can gather data on it." It's perfect for forcing spend to untested assets to see how they perform.

You've been using a scaling tool (CBO) for a testing job, which is why it's not working. You're asking the screwdriver to hammer in a nail. The solution isn't to throw away the screwdriver, it's to build a proper toolbox and use the right tool for the right job.

What you need is a proper funnel structure. This is how you move away from just throwing random creatives at a broad audience and start speaking to potential customers in a way that makes sense based on their relationship with your brand. For most eCommerce businesses, we build a structure with three core campaigns running at all times:

1. Top of Funnel (ToFu) - Prospecting: This is where you find new customers. People who have never heard of you before. The goal here is to introduce your brand and products to cold audiences based on their interests, demographics, or behaviours. This is where the majority of your budget will go.
2. Middle of Funnel (MoFu) - Retargeting: This campaign targets people who have shown some interest but haven't made a purchase yet. They might have visited your website, watched one of your videos, or engaged with an Instagram post. Here, the message changes from "Hey, we exist!" to "Hey, remember us? Here's why you should come back."
3. Bottom of Funnel (BoFu) - Re-engagement: This is for the hottest audiences. People who added a product to their cart but didn't buy, or initiated checkout. It can also be used to target past customers to encourage repeat purchases. The message here is very direct, often with an incentive like free shipping or a small discount to get them over the line.

Building your account this way organises everything. It stops you from lumping cold audiences and hot retargeting audiences into the same CBO campaign, which is a massive mistake. CBO will almost always favour the 'easy' conversions from the hot BoFu audience, starving your ToFu ad sets of the budget they need to find new customers. Seperating them is non-negotiable for scaling.

You probably should prioritise your audience targeting...

Right, now that we have a structure, we can talk about the most importent part of any Meta ads strategy: the audience. People get obsessed with creatives, but I promise you, an average creative shown to a perfect audience will beat a masterpiece creative shown to the wrong audience every single time.

Your job isn't to find people who *might* be interested in your product. Your job is to find the people who have the specific, urgent problem that your product solves, and who are ready to spend money to fix it. This means you need to be strategic about who you target, especially at the Top of Funnel.

Within your ToFu campaign, you're going to test different types of cold audiences. But there's an order of priority. Some audiences are inherently higher quality than others. Here’s how I’d prioritise them:

1. High-Intent Lookalike Audiences: This is your goldmine. You create a "Lookalike" audience from a list of your best customers—people who have bought multiple times or have the highest average order value. You're telling Meta, "Here are my best customers, go and find me a million more people who look and behave exactly like them." This is almost always the best-performing cold audience. If you have at least 100-200 past purchasers, this is where you start. You can then work your way down the funnel, creating lookalikes of people who added to cart, then lookalikes of website visitors. The closer the source audience is to a purchase, the better the lookalike will be.
2. Detailed Targeting (Interests & Behaviours): This is what most people start with. It can work well, but it needs to be done thoughtfully. The mistake I see all the time is targeting interests that are too broad. Let's say you sell high-end coffee beans. Targeting the interest "Coffee" is a terrible idea. You'll hit everyone from people who drink instant coffee once a week to genuine connoisseurs. Instead, you'd target interests like "James Hoffmann," "Aeropress," "Speciality Coffee Association," or competitor brands like "Square Mile Coffee Roasters." You need to target the niche signals that only your true ideal customer would exhibit. Think about what podcasts they listen to, what magazines they read, what tools they use, what influencers they follow.
3. Broad Targeting: This means you literally target an entire country with maybe just an age and gender filter. You rely entirely on the Meta pixel data and algorithm to find the right people. This can work incredibly well, but only *after* your pixel has thousands of purchase events. For a new or smaller account, going broad is like setting your money on fire. You should only test this once you've scaled to a much higher spend and have a tonne of conversion data.

So, inside your ToFu prospecting campaign, you'd set up multiple ad sets. One for your "1% Purchasers Lookalike," another for a group of "Niche Coffee Influencer" interests, another for "Competitor Brand" interests, and so on. By separating them, you can clearly see which audience is performing best, and allocate more budget accordingly.

You'll need a proper creative testing framework...

Now we can finally talk about testing those lovely AI creatives you're making. You've got the right idea that creative testing is the engine of growth on Meta. But you've been trying to service that engine while the car is speeding down the motorway. It's messy and ineffective. We need to build a dedicated garage for it.

This is what we do: we create a completely separate "Testing Campaign". This campaign's only job is to find winning creatives. Here's how it's structured:

• Campaign Setup: Set this campaign to ABO (Ad Set Budget Optimisation). This is critical. We want to *force* the budget to be spent evenly across our new creatives so we can gather data on all of them, not just the one the algorithm likes first.
• Budget: You don't need a huge budget. Maybe 10-20% of your total daily spend. So in your case, around £70-£140 per day for the testing campaign.
• Ad Set Setup: Inside this campaign, create one ad set. For the audience, use your single best-performing cold audience. This could be your 1% purchasers lookalike, or your best interest-based audience. We want to test our creatives against a reliable, proven audience to reduce variables.
• Creative Setup: Inside this single ad set, you'll place your new creatives to test. Don't do three every day. Start with a batch of maybe 3-5 new creatives at the start of the week. Let them run for at least 3-4 days, or until each ad has spent at least 1-2x your target Cost Per Purchase.

After 3-4 days, you analyse the results. You're looking for the "winners". A winner isn't just about ROAS. You're looking at a combination of metrics: a high Click-Through Rate (CTR) shows the ad is engaging, a low Cost Per Click (CPC) shows it's efficient, and of course a low Cost Per Purchase (CPA) and high ROAS show it's effective. Any ad that clearly outperforms the others is a winner.

What do you do with a winner? You turn it off in the testing campaign and you move it into your main "Scaling Campaign" (your ToFu CBO campaign). Now, your main scaling campaign is only ever filled with creatives that have already proven themselves. This makes your main CBO campaign incredibly stable and efficient, because it's only ever optimising its budget between known winners. The testing campaign is your lab; the scaling campaign is your factory.

This systematic approach ends the chaos. No more daily uploads. No more creatives getting £0 spend. It's a clear, repeatable process: Test in ABO -> Find Winners -> Move Winners to CBO for Scaling.

Let's talk about the numbers that actually matter...

You mentioned your ROAS is around 3.5. As I said, that's a good starting point. But "good" is relative. A 3.5x ROAS could be fantastic for a business with 80% profit margins, but it could be losing money for a business with 30% margins. The metric that truly matters, the one that should guide every decision you make, is the ratio between your Customer Lifetime Value (LTV) and your Customer Acquisition Cost (CAC).

LTV is the total profit you expect to make from a single customer over the entire duration of their relationship with you. CAC is simply what you pay to acquire that customer (your CPA from the ads). A healthy, scalable business typically aims for an LTV:CAC ratio of at least 3:1. This means for every £1 you spend to get a customer, you get at least £3 back in profit over their lifetime.

Knowing your LTV is a superpower. It tells you exactly how much you can afford to spend to acquire a customer. If you know your average customer is worth £300 in profit to you, then you know you can afford to spend up to £100 to acquire them. This frees you from the tyranny of worrying about a low ROAS on a particular day. You can confidently spend on campaigns knowing that even if the immediate return is only 2x, the long-term return will be profitable.

Calculating it can seem complex, but we can make a pretty good estimate with three simple numbers:

1. Average Revenue Per Account (ARPA): How much does an average customer spend with you per month (or year, depending on your business model)?
2. Gross Margin %: What is your profit margin on that revenue? (Revenue - Cost of Goods Sold) / Revenue.
3. Monthly Churn Rate %: What percentage of your customers do you lose each month?

The formula is: LTV = (ARPA * Gross Margin %) / Monthly Churn Rate

Let's play with some numbers. Use the calculator below to get a feel for your own LTV. It might surprise you how much you can actually afford to spend per lead.

Once you know this number, you have your North Star. You can make much smarter decisions, weather the daily ups and downs of the algorithm, and focus on acquiring high-value customers, not just cheap clicks.

This is the main advice I have for you:

To pull all this together, here’s a summary of the actionable steps I’d recommend you take to get your account back on track and ready for sustainable growth. This isn't about quick fixes; it's about building a professional, robust system.

Implementing a system like this is a significant shift from what you're currently doing. It requires more strategic thinking upfront but leads to far less stress and much more predictable results in the long run. The goal is to move from being a reactive 'ad-launcher' to a proactive 'system-builder'.

It can be a lot to take on by yourself, especially when you're also trying to run your business. This is often where getting some expert help can make a huge difference, not just in getting better results, but in saving you time and preventing costly mistakes. We've untangled and scaled accounts just like yours many times over. I remember one eCommerce client in the women's apparel space who was stuck in a similar loop, and by implementing this exact testing and scaling structure, we helped them achieve a 691% return on their ad spend.

If you’d like to have a more detailed chat, we offer a free, no-obligation initial consultation where we can go through your ad account together on a call and map out a specific plan for you. It's a great way to get a second pair of expert eyes on your strategy.

Regards,

Team @ Lukas Holschuh